Getting started

This library relies on the tch crate for bindings to the C++ Libtorch API. A more detailed installation of this library can be found at the link to this repository.

Inference

``` use rusty_yolo; use tch;

fn main() { let device = tch::Device::cudaifavailable(); let yolomodel = rustyyolo::YOLO::new("model.pt", 384, 640, device); let mut original_image = tch::vision::image::load("/images/zidane.jpg").unwrap();

let results = yolo_model.predict(&original_image, 0.25, 0.35);
yolo_model.draw_rectangle(&mut original_image, &results);
tch::vision::image::save(&original_image, "images/result.jpg").unwrap();

}

```